Add cache-dir argument to cloner

Clone repos from cache-dir if they exist, then reset their origins
to be the git base url and update from there.

Change-Id: Ic42961aa5da9e4a66524ec14758d421944403f91

4 files changed, 94 insertions, 3 deletions

diff --git a/doc/source/cloner.rst b/doc/source/cloner.rst
index bb33f828e..2ddf0b59c 100644
--- a/doc/source/cloner.rst
+++ b/doc/source/cloner.rst
@@ -75,3 +75,15 @@ cloner processes the clones in the order supplied, so you should swap the
 projects::
 
   zuul-cloner project project/plugins/plugin1
+
+Cached repositories
+-------------------
+
+The ``--cache-dir`` option can be used to reduce network traffic by
+cloning from a local repository which may not be up to date.
+
+If the ``--cache-dir`` option is supplied, zuul-cloner will start by
+cloning any projects it processes from those found in that directory.
+The URL of origin remote of the resulting clone will be reset to use
+the ``git_base_url`` and then the remote will be updated so that the
+repository has all the information in the upstream repository.
diff --git a/tests/test_cloner.py b/tests/test_cloner.py
index 166c48666..ab2683d81 100644
--- a/tests/test_cloner.py
+++ b/tests/test_cloner.py
@@ -59,6 +59,64 @@ class TestCloner(ZuulTestCase):
                 os.path.join(self.upstream_root, project))
         return repos
 
+    def test_cache_dir(self):
+        projects = ['org/project1', 'org/project2']
+        cache_root = os.path.join(self.test_root, "cache")
+        for project in projects:
+            upstream_repo_path = os.path.join(self.upstream_root, project)
+            cache_repo_path = os.path.join(cache_root, project)
+            git.Repo.clone_from(upstream_repo_path, cache_repo_path)
+
+        self.worker.hold_jobs_in_build = True
+        A = self.fake_gerrit.addFakeChange('org/project1', 'master', 'A')
+        A.addApproval('CRVW', 2)
+        self.fake_gerrit.addEvent(A.addApproval('APRV', 1))
+
+        self.waitUntilSettled()
+
+        self.assertEquals(1, len(self.builds), "One build is running")
+
+        B = self.fake_gerrit.addFakeChange('org/project2', 'master', 'B')
+        B.setMerged()
+
+        upstream = self.getUpstreamRepos(projects)
+        states = [
+            {'org/project1': self.builds[0].parameters['ZUUL_COMMIT'],
+             'org/project2': str(upstream['org/project2'].commit('master')),
+             },
+            ]
+
+        for number, build in enumerate(self.builds):
+            self.log.debug("Build parameters: %s", build.parameters)
+            cloner = zuul.lib.cloner.Cloner(
+                git_base_url=self.upstream_root,
+                projects=projects,
+                workspace=self.workspace_root,
+                zuul_branch=build.parameters['ZUUL_BRANCH'],
+                zuul_ref=build.parameters['ZUUL_REF'],
+                zuul_url=self.git_root,
+                cache_dir=cache_root,
+                )
+            cloner.execute()
+            work = self.getWorkspaceRepos(projects)
+            state = states[number]
+
+            for project in projects:
+                self.assertEquals(state[project],
+                                  str(work[project].commit('HEAD')),
+                                  'Project %s commit for build %s should '
+                                  'be correct' % (project, number))
+
+        work = self.getWorkspaceRepos(projects)
+        upstream_repo_path = os.path.join(self.upstream_root, 'org/project1')
+        self.assertEquals(work['org/project1'].remotes.origin.url,
+                          upstream_repo_path,
+                          'workspace repo origin should be upstream, not cache')
+
+        self.worker.hold_jobs_in_build = False
+        self.worker.release()
+        self.waitUntilSettled()
+
     def test_one_branch(self):
         self.worker.hold_jobs_in_build = True
 
diff --git a/zuul/cmd/cloner.py b/zuul/cmd/cloner.py
index 2fcaacd29..a895f2433 100755
--- a/zuul/cmd/cloner.py
+++ b/zuul/cmd/cloner.py
@@ -54,6 +54,10 @@ class Cloner(zuul.cmd.ZuulApp):
         parser.add_argument('--version', dest='version', action='version',
                             version=self._get_version(),
                             help='show zuul version')
+        parser.add_argument('--cache-dir', dest='cache_dir',
+                            help=('a directory that holds cached copies of '
+                                  'repos from which to make an initial clone.'
+                                  ))
         parser.add_argument('git_base_url',
                             help='reference repo to clone from')
         parser.add_argument('projects', nargs='+',
@@ -142,6 +146,7 @@ class Cloner(zuul.cmd.ZuulApp):
             branch=self.args.branch,
             clone_map_file=self.args.clone_map_file,
             project_branches=project_branches,
+            cache_dir=self.args.cache_dir,
         )
         cloner.execute()
 
diff --git a/zuul/lib/cloner.py b/zuul/lib/cloner.py
index 579b9c7e1..89ebada0e 100644
--- a/zuul/lib/cloner.py
+++ b/zuul/lib/cloner.py
@@ -29,13 +29,14 @@ class Cloner(object):
 
     def __init__(self, git_base_url, projects, workspace, zuul_branch,
                  zuul_ref, zuul_url, branch=None, clone_map_file=None,
-                 project_branches=None):
+                 project_branches=None, cache_dir=None):
 
         self.clone_map = []
         self.dests = None
 
         self.branch = branch
         self.git_url = git_base_url
+        self.cache_dir = cache_dir
         self.projects = projects
         self.workspace = workspace
         self.zuul_branch = zuul_branch
@@ -66,9 +67,24 @@ class Cloner(object):
         self.log.info("Prepared all repositories")
 
     def cloneUpstream(self, project, dest):
+        # Check for a cached git repo first
+        git_cache = '%s/%s' % (self.cache_dir, project)
         git_upstream = '%s/%s' % (self.git_url, project)
-        self.log.info("Creating repo %s from upstream %s",
-                      project, git_upstream)
+        if (self.cache_dir and
+            os.path.exists(git_cache) and
+            not os.path.exists(dest)):
+            # file:// tells git not to hard-link across repos
+            git_cache = 'file://%s' % git_cache
+            self.log.info("Creating repo %s from cache %s",
+                          project, git_cache)
+            new_repo = git.Repo.clone_from(git_cache, dest)
+            self.log.info("Updating origin remote in repo %s to %s",
+                          project, git_upstream)
+            origin = new_repo.remotes.origin.config_writer.set(
+                'url', git_upstream)
+        else:
+            self.log.info("Creating repo %s from upstream %s",
+                          project, git_upstream)
         repo = Repo(
             remote=git_upstream,
             local=dest,